Hi everybody,
I am following the NDK 2.1 documentation for Planned Failover:
Configuring and Running Planned Failover
In step 3, Configure the migrate-job YAML, the sample YAML contains this placeholder:
containers:
- name: ndk-dr-container
image: # <image-repo-url>:<tag>
I am not sure what image URL should be used here for the NDK recipe job.
My environment is non-air-gapped. I already created the Docker registry secret using the Nutanix Docker Hub access token:
kubectl -n ndk-recipes create secret docker-registry ndk-recipe-pull-secret \
--docker-server=https://index.docker.io/v1/ \
--docker-username=nutanixndk \
--docker-password="<REDACTED_ACCESS_TOKEN>" \
--docker-email="<my-email>" \
--save-config \
--dry-run=client -o yaml | kubectl apply -f -
The secret was created successfully:
NAME TYPE DATA
ndk-recipe-pull-secret kubernetes.io/dockerconfigjson 1
I tried using this image:
image: ndk/recipes:2.1.0
But the pod failed with the following error:
Failed to pull image "ndk/recipes:2.1.0":
failed to pull and unpack image "docker.io/ndk/recipes:2.1.0":
failed to resolve reference "docker.io/ndk/recipes:2.1.0":
pull access denied, repository does not exist or may require authorization:
server message: insufficient_scope: authorization failed
Could you please confirm the correct image URL and tag that should be used in migrate-job.yaml for NDK 2.1 Planned Failover in a non-air-gapped environment?
Also, is ndk/recipes:2.1.0 only the local image name used after loading ndk-2.1.0-recipes-img.tar from the air-gapped bundle, or is it supposed to be pullable from Docker Hub?
Thanks.
